CCI-Data.py

import json

import datasets
import traceback
import os

logger = datasets.logging.get_logger(__name__)


_DESCRIPTION = "CCI is a Chinese webtext dataset."

_URL_LISTS = {
    "webtext": "urls/webtext.txt",
}

class CCIDataConfig(datasets.BuilderConfig):

    def __init__(self, *args, subsets, **kwargs):
        super(CCIDataConfig, self).__init__(**kwargs)
        self.subsets = subsets


class CCIData(datasets.GeneratorBasedBuilder):

    BUILDER_CONFIGS = [
        CCIDataConfig(
            name = 'default',
            subsets = list(_URL_LISTS.keys()),
            version=datasets.Version("1.0.0", ""),
            description="CCI",
        ),

        CCIDataConfig(
            name = 'webtext',
            subsets = ['webtext'],
            version=datasets.Version("1.0.0", ""),
            description="CCI webtext subset",
        )
    ]

    DEFAULT_CONFIG_NAME = "webtext"

    def _info(self):
        return datasets.DatasetInfo(
            description=_DESCRIPTION,
            features=datasets.Features(
                {
                    "id": datasets.Value("string"),
                    "content": datasets.Value("string")
                }
            ),
            supervised_keys=None,
        )

    def _split_generators(self, dl_manager):
        url_lists = dl_manager.download_and_extract({
            subset: _URL_LISTS[subset] for subset in self.config.subsets
        })

        urls = {}

        for subset, url_list in url_lists.items():
            with open(url_list, encoding="utf-8") as f:
                urls[subset] = [line.strip() for line in f]

        downloaded_files = dl_manager.download(urls)

        return [
            datasets.SplitGenerator(
                name=datasets.Split.TRAIN,
                gen_kwargs = {
                    "files": {
                        subset: downloaded_files[subset]
                        for subset in self.config.subsets
                    }
                }
            )
        ]

    def _generate_examples(self, files):
        key = 0
        for subset in files:
            for path in files[subset]:
                with open(path, encoding="utf-8") as f:
                    for i, row in enumerate(f):
                        try:
                            data = json.loads(row)
                            yield key, {
                                    "id": data["id"],
                                    "content": data["content"]
                                }
                            key += 1
                        except Exception as e:
                            print(f'Subset: {subset}')
                            print(f'Path: {path}')
                            print(f'Row: {row}')
                            traceback.print_exc()

                            raise e